N221-014 TITLE: Synthetic Aperture Radar High Resolution Imaging when Performing Random Nonrepeating Radar Orbits

OUSD (R&E) MODERNIZATION PRIORITY: General Warfighting Requirements (GWR);Networked C3

TECHNOLOGY AREA(S): Air Platforms;Battlespace Environments;Information Systems

OBJECTIVE: Develop innovative Synthetic Aperture Radar (SAR) image formation/detections techniques for aerial vehicles performing Coherent Change Detection (CCD) that permits randomized radar orbits.

DESCRIPTION: Modern synthetic aperture radar signal processing algorithms retrieve accurate and subtle information regarding a scene that is being interrogated by an airborne radar. An important application of "continuous-stare" SAR systems involves detecting changes in an imaged scene. Current CCD radar techniques require flying the same orbit path repeatedly in order to look for changes in the scene. While this is acceptable in a benign threat environment, this predictable flight profile will be lethal to the air vehicle in a high-threat environment.

Achieving the alignment necessary between images (especially for CCD) is often difficult due to many factors beyond the control of the platform and sensor, including imaging geometry issues, air vehicle motion, errors in motion compensation, difficult clutter environments, clutter motion, and meteorological issues. Even when good alignment is obtained, weaker stationary target signature may be overcome by the surrounding clutter or masked by false alarms, requiring more sophisticated alignment algorithms and change metrics to extract the relevant image change information. Also, the steep depression angles required for urban imagery aggravate the effects of mismatched imaging geometry on change detection. These effects can be considerable, especially for CCD. Noncoherent change detection (NCCD) is often difficult in urban areas, for example, because large cultural object scatterers and their side lobes may be difficult to align (especially when imaging geometries are different) and may overwhelm weaker stationary target signatures. High-frequency CCD is potentially capable of detecting extremely subtle terrain disturbances, but is even more sensitive to alignment issues, typically producing an overwhelming number of naturally occurring false alarms, even in relatively benign cases. Inability to perform change detection (CD) may result in missed opportunities for extraction of significant information.

This SBIR topic seeks to develop and demonstrate techniques and transforms enabling CCD to be conducted with orbits that are randomized and nonrepeating. These techniques and transforms will be applicable to any aerial vehicle that is conducting SAR CCD missions under degraded conditions and various deployment environments. These techniques and transforms will be able to compare the coherent and/or noncoherent reference and test SAR images: (a) to detect image changes in randomized CCD radar orbits; (b) to extract automatic features such as stationary-vehicle Doppler "smears" embedded in urban clutter and other alternative change detection metrics; (c) sensitivity to various image formation techniques, including wavefront reconstruction (WR) and the polar format algorithm (PFA); and (d) sensitivity to phase history processing and conditioning methods.

Developed algorithms, techniques, transforms, and a simulation tool to estimate the SAR performance producing high-resolution radar imagery of stationary objects being performed by various aerial vehicles performing randomized CCD radar orbits will be tested during one, or possibly two, Government Rapid Prototype Experimentation Demonstration (RPED).

The offeror’s proposal must clearly explain how an aerial vehicle flight path variation SAR imagery collection will be accounted for with respect to:

1. Spatial registration of the reference SAR image with respect to the test SAR image using the available air vehicle platform motion data (e.g., Global Positioning System (GPS), Inertial Measurement Unit (IMU), etc.).
2. Spatially varying motion compensation (on points on ground plane and elevation) in a three-dimensional spatial domain using GPS denied navigation filtering software and simulation to assess SAR randomized and nonrepeating orbits imagery quality when the GPS is available or denied.
3. Spectral registration of both the test SAR image and the georegistered reference SAR image to extract the common Doppler data in the two images using the available air vehicle platform motion data.
4. Blind calibration of variations of the Image Point Responses (IPR) of the resultant (spatially and spectrally registered) reference and test SAR images using applicable adaptive filtering methods.

Air vehicles equipped with multichannel along-track monopulse/displaced phase center antenna-SAR (ATM-SAR) and high-Pulse Repetition Frequency AzScan-SAR Doppler Beam Sharpening-SAR (DBS-SAR) may be included in but are not required to be part of this SBIR topic.

Air vehicles performing Ground Moving Target Indicator (GMTI) using SAR imagery are not to be considered and are not part of this topic.

PHASE I: Develop and demonstrate techniques and transforms that exploit fundamental mathematical and physical properties of SAR signals to detect changes in imagery of a scene that are acquired via randomized CCD radar orbits of a SAR aerial platform using a single channel radio frequency (RF) sensor. Develop and provide techniques and transforms to retrieve the common spatial spectral information in randomized and nonrepeating orbits SAR imagery for coherent and noncoherent change detection (NCD). Develop and provide signal models for the effects of measured randomized and nonrepeating orbits SAR data calibration errors as related to radar electronics phase errors, and unknown motion errors. Deliver an analytical study of the randomized and nonrepeating orbits SAR signal, and identify its information contents in the spatial and spectral domains. The Phase I effort will include prototype plans to be developed under Phase II.

PHASE II: Further study, develop, and improve analytical principles developed in Phase I. Validate and mature the mathematical modeling and processing trade-analysis using an Integrated Fly-out Simulation (IFS) testbed to exploit randomized and nonrepeating orbits SAR imagery. Demonstrate, with minimal additional data processing in the image formation process, in a relevant flight environment the developed signal processing techniques and transforms that exploit randomized and nonrepeating single-channel SAR data acquired at different time points for high-resolution imagery of stationary objects during a Government sponsored RPED. Furthermore, real-time simultaneous imaging, as well as CCD/NCD, are to be demonstrated while the SAR data is being collected over a full aperture (slow-time).

PHASE III DUAL USE APPLICATIONS: Further research and development will be directed toward refining final Synthetic Aperture Radar (SAR) image formation/detections techniques. Incorporate these techniques based on results from tests conducted during Phase II.

Deploy Synthetic Aperture Radar Image (SAR) formation/detections techniques, in relevant environment aerial test environments, to validate techniques. Document lessons learned (what worked, what did not, areas of improvement). Identify gaps in SAR image formation/detections techniques and propose a solution to the identified gap to the Government working groups.

The completion of this phase would result in a mature capability, which would undergo an appropriate operational demonstration, such as surveillance and reconnaissance. These SAR image formation/detections techniques should prove the ability to provide high-resolution imagery of stationary objects by various aerial vehicles performing randomized CCD radar orbits.

Continue relationships with radar manufacturers with the objective of placing these Synthetic Aperture Radar Image (SAR) formation/detections techniques in major defense and commercial radars.

From a military application, these Synthetic Aperture Radar I (SAR) image formation/detections techniques would enable hypersonic air vehicles to have high-speed radar target detection, identification, and discrimination capability of stationary objects. From a commercial application, homeland security and commercial applications include guidance and control for robotic systems used in hazardous environments, and materials handling applications involving cranes; and loading equipment, and industrial equipment used in assembly, welding, inspection, and other similar operations. These algorithms could be used to support commercial ground mapping applications and current radar system performance for border patrol, drug traffic monitoring, perimeter surveillance, and air traffic control applications.

REFERENCES:

1. Ranney, K. I., & Soumekh, M. (2005). Signal subspace change detection in averaged multilook SAR imagery. IEEE Transactions on Geoscience and Remote Sensing, 44(1), 201-213. https://doi.org/10.1109/TGRS.2005.859956.
2. Ranney, K., & Soumekh, M. (2005, May). Adaptive change detection in coherent and noncoherent SAR imagery. In IEEE International Radar Conference, 2005. (pp. 195-200). IEEE. https://doi.org/10.1109/RADAR.2005.1435818.
3. Burns, B., Clark, W., Alexander, J., Soumekh, M., Dorff G., Plaskyc, B., & Moussally, G. (2008, June). Change detection with delta-heading dual-pass UWB wide-beamwidth airborne SAR. Proceedings of Tri-Service Radar Symposium. https://www.mssconferences.org.
4. Himed, B., & Soumekh, M. (2006). Synthetic aperture radar–moving target indicator processing of multi-channel airborne radar measurement data. IEE Proceedings-Radar, Sonar and Navigation, 153(6), 532-543. https://doi.org/10.1049/ip-rsn:20050128.
5. Melvin, W. L., Wicks, M. C., & Brown, R. D. (1996, May). Assessment of multichannel airborne radar measurements for analysis and design of space-time processing architectures and algorithms. In Proceedings of the 1996 IEEE National Radar Conference (pp. 130-135). IEEE. https://doi.org/10.1109/NRC.1996.510669.
6. Melvin, W. L., & Wicks, M. C. (1997, May). Improving practical space-time adaptive radar. In Proceedings of the 1997 IEEE National Radar Conference (pp. 48-53). IEEE. https://doi.org/10.1109/NRC.1997.588124.
7. Himed, B., Salama, Y., & Michels, J. H. (2000, May). Improved detection of close proximity targets using two-step NHD. In Record of the IEEE 2000 International Radar Conference [Cat. No. 00CH37037] (pp. 781-786). IEEE. https://doi.org/10.1109/RADAR.2000.851934.
8. Soumekh, M. (1999). Synthetic aperture radar signal processing with matlab algorithms (Vol. 7). New York: Wiley. https://www.worldcat.org/title/synthetic-aperture-radar-signal-processing-with-matlab-algorithms/oclc/39458891&referer=brief_results.
9. Melvin, W. L. (1996). RL-TM-96-5: Sample Selection for Covariance Estimation in Practical Airborne Environments. Rome Lab Rome NY. https://apps.dtic.mil/sti/pdfs/ADA316361.pdf.
10. Sanyal, P. K., Melvin, W. L., & Wicks, M. C. (1997, June). SENSIAC-SENS-TSRS-1997-25: Space-time adaptive processing for advanced airborne surveillance (AAS) bistatic radar (U). Proceedings of the 43rd Annual Tri-Service Radar Symposium, 1997 (Vol. 1). https://www.mssconferences.org.
11. Melvin, W. L. (2002, September 16–17). ADP014046: Application of STAP in advanced sensor systems [Conference paper]. RTO SET Lecture Series, Istanbul, Turkey. https://apps.dtic.mil/sti/citations/ADP014046.
12. Himed, B., & Soumekh, M. (2005, December). AFRL-SN-RS-TR-2005-388: Synthetic Aperture Radar-Moving Target Indication (SAR-MTI) Processing of Multi-Channel Airborne Radar Measurement (MCARM) Data. Rome Lab Rome NY. https://apps.dtic.mil/sti/citations/ADA443155.
13. Melvin, W. L., Callahan, M. J., & Wicks, M. C. (2002, April). Bistatic STAP: application to airborne radar. In Proceedings of the 2002 IEEE Radar Conference (IEEE Cat. No. 02CH37322) (pp. 1-7). IEEE. https://doi.org/10.1109/NRC.2002.999683.

KEYWORDS: Radar; SAR; imagery; detection; orbit; target